Applied AI and Human Collaboration

Research on how AI systems are deployed in real-world professional, creative, and educational contexts. Covers human-AI co-writing, domain-specific applications, and the social and cognitive dynamics that emerge when language models interact with users.

503 notes (primary) · 327 papers · 6 sub-topics

View as

Expertise in the Age of AI Content

81 notes

Did ChatGPT cause Stack Overflow posting to decline?

Researchers used a difference-in-differences model to test whether public programming Q&A posting fell after ChatGPT's release. This matters because it could signal whether AI tools are shifting knowledge from public commons to private use.

Do authorship labels change how AI judges evaluate rule violations?

When AI evaluators see a constraint-breaking text, does knowing whether a human or AI wrote it shift their judgment? This tests whether AI judges apply consistent standards or defer to human authority.

Can AI skills help older or less-educated job candidates?

Do AI certifications and skills reduce hiring penalties faced by older workers or those without bachelor's degrees? This matters because it tests whether AI upskilling could level the job market for disadvantaged groups.

Do AI skills help candidates get more job interviews?

Explores whether recruiters treat AI skills as a valuable hiring signal and how credentials compare to self-declared proficiency in shaping interview invitations.

Do AI slop accusations actually detect AI text?

When online communities label comments as AI-generated slop, are they identifying genuine machine writing or enforcing social boundaries? This asks whether the accusation register tracks real detection or functions as gatekeeping.

What dimensions make text feel like AI slop?

Can we break down the vague notion of AI slop into measurable components? Researchers coded expert definitions to find which specific text properties people associate with low-quality generated writing.

Do AI writing assistants push non-Western writers toward Western styles?

This experiment tests whether GPT-4o autocomplete nudges Indian writers away from their native writing conventions while giving American writers larger productivity gains, raising questions about whose norms AI systems encode.

Can AI creators match human creators through posting volume alone?

On a Chinese short-video platform, do AI-generated content creators achieve comparable engagement totals to human creators by uploading significantly more videos, despite lower per-video consumer preference?

Does receiving AI-written work change how we judge the sender?

When recipients receive polished AI-generated work, do they form lower opinions of the sender's creativity, capability, and trustworthiness? Understanding this perception gap matters for how AI-mediated collaboration affects professional relationships.

How much work that employees receive is actually unhelpful AI content?

A 2025 survey asked U.S. desk workers to estimate what share of their received work consists of low-quality AI output. Understanding this helps measure whether AI tools are creating friction rather than efficiency in knowledge work.

Do platforms inevitably decline through value extraction cycles?

Does Doctorow's enshittification model describe a predictable platform lifecycle, or merely illustrate selective cases? The question matters because it determines whether platform decay is structural or contingent.

Does TikTok use special boosts to inflate partner videos?

Forbes reported that TikTok employees use an internal "heating tool" to selectively amplify videos from accounts the platform wants as business partners. Understanding whether this practice exists and how it works matters for evaluating platform fairness and creator dependence.

Why did AI article share stop growing after 2025?

Graphite reports a plateau in AI-generated articles at 50% but cannot distinguish whether search engines penalize AI content, detection tools are missing more AI text, or both. The excerpt leaves this critical cause untested.

Has AI-generated content stopped growing on the web?

After ChatGPT's launch, the share of AI-generated articles rose sharply then plateaued near 50% by early 2025. The question explores whether this plateau reflects market saturation, search penalties, or detector limits.

Do hiring managers and job seekers agree on AI fairness?

Explores the gap between how hiring managers and job seekers perceive AI's role in hiring decisions. Understanding this disagreement matters because it reveals whether AI adoption is actually improving fairness or eroding trust.

Are job applicants and employers locked in an escalating AI arms race?

This explores whether applicant use of AI tools to game applications and employer AI filtering systems are feeding each other in a self-reinforcing cycle, and whether evidence supports this claimed loop.

Do rewrites that hide authorship also fool AI detectors?

A paper claims heavily rewritten AI text becomes both unattributable to humans and undetectable as AI-assisted, but tests only the attribution half. Does rewriting actually evade detection systems?

How much does AI rewriting erase distinctive author voice?

Does heavy AI rewriting weaken the computational signals that identify individual authors? The question matters because it bears on whether AI assistance erases stylistic distinctiveness—a possible cost of polish and consistency.

Is higher AI use by Indian writers a confound to control?

Should researchers treat differential AI reliance across cultural groups as a statistical confound to remove, or as a meaningful cultural finding about trust and technology adoption that reveals homogenization effects?

How often does LinkedIn wrongly flag legitimate posts?

LinkedIn claims 94 percent accuracy on detecting AI-generated content, but hasn't released independently verified data. The real question is how many legitimate writers get quietly demoted by false positives.

Do Kaggle medals still predict performance after AI arrived?

This research asks whether Kaggle's medal credentials retained their ability to forecast actual performance as generative AI transformed the platform. It matters because it tests whether verified credentials stay meaningful when the tools behind them change.

Can hiring signals survive when AI makes cover letters worthless?

As AI undermines the cover letter's ability to signal candidate quality and interest, what alternative signals might employers rely on instead? This matters because hiring depends on cutting through noise to find good matches.

Does AI-generated cover letter access weaken hiring signals?

When job platforms give candidates AI tools to write cover letters, does the quality of letters stop predicting who gets hired? This matters because hiring relies on signals to identify strong fits.

Do LLM raters show hidden demographic preferences that disclosure erases?

Explores whether language models systematically favor certain demographic groups when their AI involvement is not disclosed, and whether that preference disappears under transparency. This matters because it reveals potential fragility in AI alignment training.

How fast did LLM writing adoption actually spread?

Does LLM-assisted writing use follow a predictable adoption curve across different sectors? Understanding the speed and pattern of adoption helps explain how quickly new AI tools reshape professional communication.

Does LinkedIn's AI detection actually improve conversation quality?

LinkedIn claims its AI-flagged content filter preserves human conversation by limiting reach of generic posts. But the 94% accuracy figure is unverified, and the impact on legitimate writers remains unmeasured.

Will LinkedIn's AI slop flag reduce AI-generated posts?

LinkedIn introduced a user flag for AI-generated content and says it will cut distribution of generic AI posts. But does the flag actually change what members see, and will it shift the measured share of AI content on the platform?

Are recruiters and job seekers really adopting AI in hiring?

LinkedIn reports that 93% of recruiters and 81% of job seekers plan to use or are using AI in hiring. But how were these figures gathered, and do they reflect actual behavior or stated intentions?

Does LinkedIn's AI slop button actually reduce low-quality content?

LinkedIn introduced a report button for AI-generated posts to train classifiers that filter recommendations. But the announcement provides no accuracy data, false-positive rates, or evidence the system works.

Does LinkedIn's generic content filter actually work fairly?

LinkedIn claims its system identifies generic AI-like posts 94% of the time and limits their spread. But the company hasn't published its false-positive rate or defined what makes content generic, leaving open whether human writers get caught in the filter.

Do LinkedIn's AI hiring tools actually produce better hires?

LinkedIn reports that its AI screening tools save recruiters time and help discover new candidates. But the company has not published data on whether candidates found through AI screening become better employees, stay longer, or perform better once hired.

Do private AI flags actually change how people write?

LinkedIn flags posts with heavy AI use privately to help writers refine their work, but the announcement includes no evidence that the flags change writing behavior, disclosure, or posting rates.

Do readers engage less with AI-generated social media posts?

On Medium, posts labeled as AI-generated received fewer likes and comments than human-written posts. The question is whether this gap reflects genuine reader preference or stems from other factors like author differences or detector errors.

How much LinkedIn content is AI-generated right now?

Originality.ai's detector found 81% of sampled July 2026 LinkedIn posts marked as likely AI-written, up from roughly 50% in late 2024. Understanding this trend matters for assessing platform authenticity and user trust.

Why does Reddit's AI share seem so low compared to others?

Reddit shows a 4.4% aggregate AI share, but this hides a crucial split: most content is replies (98.1% human), while top-level posts reach 11.6% AI. Does composition explain the apparent platform difference?

Why does LinkedIn have the most AI-generated posts?

Researchers scanned over 1 million social media posts and found LinkedIn's longform content had the highest rate of AI generation. The question explores whether platform design, user behavior, or both drive this concentration.

Do people fear judgment when they use AI at work?

This research explores whether workers expect others to view them as less competent or diligent when using AI tools, and whether that fear affects their willingness to disclose tool use to managers and colleagues.

Why aren't LinkedIn users adopting AI post-writing tools?

LinkedIn's CEO suggests AI-drafted posts underperform expectations because public visibility creates reputational risk. This explores whether social penalty—being called out for machine-written content—actually suppresses adoption of platform writing tools.

Does hidden AI use cost more trust when exposed?

When AI use is discovered after being kept secret, does trust decline more steeply than if disclosed upfront? The question matters because it suggests concealment may carry hidden risks beyond the initial disclosure penalty.

Does disclosing AI use damage how trustworthy you seem?

When people learn you used AI to create work, do they trust you less? Schilke and Reimann tested this across 13 experiments with over 5,000 participants to understand whether transparency about AI reliance backfires.

Did ChatGPT displace only low-quality Stack Overflow posts?

After ChatGPT's release, Stack Overflow posts received similar vote scores, but votes are an imperfect quality measure. The research uses voting patterns to infer what type of content was displaced, though this inference remains unvalidated against expert judgment.

Does LinkedIn's 94% accuracy apply to human posts wrongly limited?

LinkedIn claims 94% accuracy identifying generic content, but the excerpt provides no false-positive rate, sample details, or comparison with human-written posts. The scope of this accuracy claim remains unclear.

Does YouTube's AI-persona rule actually prevent viewer confusion?

YouTube denies monetization to AI personas presenting as human experts on sensitive topics, citing viewer confusion as the harm. But does evidence support that confusion is the real problem, and can platforms reliably identify when this rule applies?

Does YouTube care how creators make videos or what they produce?

YouTube's monetization rules focus on the finished video's quality and originality, not the production process. This note explores whether the policy actually distinguishes AI-assisted work from human-made work in practice.

How much machine-generated text actually appears on Reddit?

Researchers ran a detector across millions of Reddit posts and comments to measure how prevalent AI-written content is on the platform. Understanding this prevalence matters for assessing Reddit's authenticity and the scale of AI adoption in online communities.

Does AI essay use hurt admissions chances despite quality gains?

This study explores whether applicants who use AI to write essays face admission penalties, even when those essays show higher writing quality. The tension matters because it suggests institutions may discount AI-assisted work regardless of its objective merit.

How much did retiring a competition format hurt medal credibility?

Kaggle phased out upload-format competitions before AI arrived, but kept displaying their medals at face value as credentials aged. How much of the decline in medal informativeness came from this institutional stranding rather than AI effects?

Do unfounded AI accusations harm human writers instead?

When readers accuse writers of using AI without evidence, does that flip who suffers epistemic injustice? This explores whether blanket distrust of suspected AI text can wrong human authors at scale.

Do admissions officers penalize essays they suspect are AI-written?

An experiment tested whether admissions officers can distinguish AI from human writing and whether suspected AI authorship affects their ratings. This matters because it could explain why AI-written essays face lower acceptance rates.

Does AI cover letter writing change what employers value?

When AI tools automate cover letter creation, do employers shift their attention to different signals about worker quality? This matters because it shows how AI automation affects the job market beyond just changing efficiency.

Do authorship labels bias how we judge literary quality?

When readers and AI systems know whether text is human- or AI-written, does that label shift their judgments of the same passage? This matters because it tests whether evaluation is based on actual content or on authorship cues.

Does cheap AI simulation break the credibility of costly signals?

If AI can now cheaply produce outputs that once required genuine mental effort, do the costly signals people rely on to prove their honesty and knowledge still work? This matters in low-trust settings where reputation cannot enforce honesty.

Does trust in unlabeled AI messages decline as awareness grows?

Researchers predict that rising public awareness of generative AI may erode the default trust readers extend to unlabeled messages, but their single-wave experiment cannot track this change over time or across populations with different AI exposure.

Is AI-generated content rising faster on some platforms?

A detector applied to 2.4M posts across Medium, Quora, and Reddit from 2022–2024 found AI attribution rates climbing sharply on two platforms but barely budging on one. Why do adoption patterns differ so dramatically?

Does disclosing AI assistance make readers trust articles less?

When articles carry a label saying they used AI tools, do human and AI raters downgrade their quality assessments? This matters because writers worry disclosure could harm how their work is received.

How does revealing AI authorship change reader trust?

When readers learn that AI wrote part of a text, do they trust the author less? This study tested whether disclosure of AI involvement shifts how readers judge an author's trustworthiness, caring, and likability across different types of writing.

Which AI design principles for social media have research support?

The paper proposes four principles for ethical AI integration on social media. But how many actually rest on tested evidence versus untested assumptions? The distinction matters for platforms deciding what to implement.

Does AI literacy reduce the damage from AI disclosure?

When readers learn that AI was used in writing, does their knowledge about AI systems affect how negatively they judge the work? Understanding this matters for writers deciding whether to disclose.

Can people reliably spot content made by AI?

This systematic review of 30 studies asks whether human judgment can distinguish AI-generated text, images, and voice from human-created content, and whether detection accuracy has improved as AI becomes more realistic.

Do language models favor resumes they rewrote themselves?

When LLM evaluators choose between resumes describing the same candidate, do they systematically prefer versions they generated over human-written originals? Testing this matters because algorithmic hiring could amplify AI-generated content at scale.

Why do readers and writers disagree on disclosure necessity?

When writers steer AI generation less intentionally, readers want more disclosure but writers want less. This reversal is puzzling—what explains why the same signal pushes the two groups in opposite directions?

Can readers tell truth from fabrication without evidence signals?

When readers see fluent text with no provenance information, do they distinguish accurate claims from AI-generated hallucinations? This tests whether presentation authority alone misleads judgment.

Why did AI tools break the effort signal in hiring?

Before LLMs, employers read proposal effort as a sign of worker ability. After AI tools became common, that signal collapsed. What changed, and does the tool itself cause it?

Do reader judgments reflect actual authorship or just their beliefs?

When readers evaluate research abstracts, do their ratings track who actually wrote them, or are they shaped by what they believe about authorship—even when those beliefs are wrong?

Can readers tell LLM abstracts from human ones?

Do readers with ML expertise reliably distinguish human-written, LLM-generated, and LLM-edited research abstracts? Understanding this matters for evaluating whether readers can serve as effective gatekeepers against LLM content.

Do readers and writers differ on AI disclosure necessity?

This vignette study explores whether readers and writers judge the necessity of disclosing AI use differently, and what conditions make disclosure feel more important to each group.

Do readers value writing authenticity they cannot detect?

Readers in Hwang et al.'s study could not distinguish AI-assisted writing from solo writing. The open question is whether readers would care about process-level authenticity if they knew about it or could perceive it.

Do readers trust unlabeled AI-written messages as much as human ones?

When AI-assisted emails lack any disclosure, do recipients judge them identically to human-written messages, or does suspicion arise even without labeling? This matters for understanding when and whether AI use needs explicit flagging.

Does cheap writing weaken hiring based on worker ability?

When AI makes written proposals cheap to produce, do employers lose their ability to identify skilled workers through application quality? This matters because applications have traditionally signaled worker talent.

Does user control over AI text shape feelings of ownership?

Explores whether giving users more influence over generated text increases their sense of authorship, and whether personalization of the AI model matters for this effect.

Which signaling equilibria does AI destruction actually harm or help?

When AI undermines costly signals that coordinate trust, it may help some people but hurt others. The research identifies this possibility but cannot yet specify which equilibria are worth preserving versus destroying.

Do LLM evaluators favor resumes written by their own model?

When the same language model drafts and evaluates resumes, does it systematically rank applicants higher if they used that model to write their application? This matters because hiring pipelines increasingly automate both resume generation and screening.

Can we judge text quality without knowing who wrote it?

Does the concept of 'slop' work as a quality judgment independent of whether a machine or human authored the text? This matters because current AI detection often conflates two separate questions: origin and quality.

Do AI writing tools improve online discussion or degrade it?

When AI assists with comments and replies, does it benefit both people writing and reading? A controlled experiment tested whether AI tools enhance or harm the quality and authenticity of online conversations.

Is the 2024 LLM writing plateau real saturation or measurement artifact?

The adoption curve for LLM-assisted writing flattened in 2024, but the cause remains unclear: either genuine saturation or models becoming too subtle to detect. Resolving this matters for understanding actual usage trends versus measurement limitations.

Does polished writing actually signal better quality work?

When evaluators judge applications and manuscripts, does rhetorical sophistication predict merit, or does it distract from verifiable evidence of competence and rigor?

Do people feel they own AI-generated text they use?

When people use personalized AI to write, do they experience a sense of authorship and ownership? Understanding this matters because it shapes whether disclosure norms around AI use are grounded in how people actually feel.

Does AI assistance actually narrow the diversity of ideas?

The paper claims AI narrows ideation diversity but only reports increased elaboration and quantity. The study lacks direct diversity measures, leaving the diversity claim unsupported by its own evidence.

Does the social penalty for AI use fade as the tool becomes ordinary?

The attribution account predicts that penalties for using unfamiliar tools should vanish once they become customary. But no longitudinal data exists on whether this actually happens with AI, leaving adoption timelines uncertain.

Does editing time on AI drafts predict hiring success?

Workers who spend more time editing AI-generated cover letters see better hiring outcomes on Freelancer.com. Understanding whether editing time reflects careful judgment, experience, or job fit could reveal what employers reward in application materials.

Where do writers locate authenticity in AI co-writing?

Do writers find authenticity in the finished text alone, or in the internal experience and process of creation? This matters for understanding what writers value when using AI tools.

Knowledge After the Web

68 notes

How is ChatGPT actually being used by real people?

A study of actual ChatGPT conversations reveals shifting patterns between work and non-work use. Understanding real usage patterns matters for predicting economic impact and technology adoption.

Could AI assistants become gatekeepers of what people see?

If Advanced AI Assistants become the default way people access information, would their intermediary position give them hidden power over what users encounter, how ideas are framed, and what options they consider?

How much are AI Overviews actually reducing organic search clicks?

Research from Ahrefs measures whether Google's AI Overviews are diverting clicks away from top-ranking pages. Understanding the scale matters for content publishers' business models.

How many Dutch voters might seek AI voting advice?

Before the 2025 Dutch election, researchers surveyed willingness to consult AI chatbots for voting guidance. Understanding who considers this and how often matters for election integrity and information quality.

Do artifact outputs reduce how critically users evaluate them?

When Claude generates code or documents as artifacts, do users provide clearer initial direction but skip fact-checking and reasoning evaluation afterward? Understanding this pattern matters for knowing whether polished outputs hide quality problems.

Does AI adoption follow wealth and mature over time?

Does Claude usage concentrate in wealthy countries, and does adoption shift from automating tasks toward augmenting human work as it deepens? Understanding this pattern matters for predicting where AI impact spreads and how its use evolves.

Can an AI agent serve both merchant and user interests fairly?

Explores whether an agent funded by merchant referral fees can make unbiased recommendations on behalf of users, or whether the payment structure creates an unavoidable conflict of interest.

How much human input did OpenAI's Navier-Stokes proof actually require?

OpenAI claimed its model produced a Navier-Stokes proof with minimal human help, but Buckmaster's account suggests the actual process involved substantial team effort, testing, and prompting. Did the public framing match what actually happened?

Will voters actually use AI chatbots for election information?

Explores how many voters plan to rely on AI chatbots versus traditional news sources for 2026 election information, and whether stated intent reflects actual behavior or real-world impact.

Does ChatGPT harm informal learning compared to Google Search?

An 8-day experiment tested whether using ChatGPT for self-directed learning produces different knowledge gains than Google Search, and explored what mechanisms might explain any differences.

Is non-human traffic really half of all internet use?

Cloudflare's network measurements show over 50% of internet traffic is now from AI crawlers and bots rather than humans. Understanding whether this reflects actual internet-wide patterns matters for how publishers, platforms, and regulators should respond.

Can an AI summary substitute for actually reading a book?

DeLong tested whether an AI-generated conspectus of a book could replace the cognitive experience of reading it. He spot-checked the output for accuracy and found it close enough to enable convincing discussion of the book's contents.

Could markets allocate scarce lab resources to AI-generated research ideas?

DeepMind researchers ask whether a market system licensing ideas to executors and paying royalties on validated results could solve the bottleneck of physical validation capacity in automated science.

Do generative UI tools actually implement their stated design rationales?

Explores whether generative UI tools build interfaces that match the design reasoning they provide. This matters because plausible-sounding rationales might persuade users to trust outputs without verification.

Do AI chatbots systematically bias voters toward extreme parties?

A Dutch regulator tested four AI chatbots for voting advice and found they consistently recommended only two parties—far-right and left-wing—while marginalizing centrist options. The question explores whether this is a design flaw or structural feature of how these tools collapse diverse inputs.

Do AI assistants reliably answer questions about news?

A major cross-country study tested whether AI assistants like ChatGPT and Gemini accurately handle news queries. Understanding this matters because many people may turn to these tools for current events.

Does AI search summaries divert traffic away from Wikipedia?

Does Google's AI Overviews feature reduce click-through traffic to Wikipedia articles by presenting synthesized answers directly in search results? This matters because it reveals whether AI intermediaries can reshape how users access information sources.

Does AI math recruitment mask the commodification of expert labor?

Whether the tech industry's recruitment of senior mathematicians for AI training represents a new form of alienated labor—where expertise is paid but understanding is erased and ownership belongs entirely to firms.

Does AI scooping force researchers to hide work in progress?

Hoel argues that AI's ability to rapidly complete half-formed ideas has destroyed the old incentive to share work-in-progress publicly, potentially driving intellectual culture underground. The question examines whether this competitive dynamic is real and widespread.

Are social media and LLMs functioning as collective hive minds?

Hoel argues that social media and LLMs already operate as hive minds—shared collective consciousnesses—rather than as separate tools. The question explores what it means to live inside such a system and whether benevolence changes its fundamental nature.

Does AI-generated slop exploit visual truth to bypass skepticism?

Horning argues that AI slop borrows the visual markers of evidence—shaky framing, documentary indexicality—to make viewers feel informed without requiring verification or belief. This raises questions about how formal resemblance to evidence can short-circuit critical judgment.

Do LLMs obscure the historical processes behind their answers?

Horning applies Lukács's reification theory to argue that LLMs present facts as fixed and given rather than as products of social process. The question is whether this theoretical framework accurately describes how LLMs shape user understanding.

Why do LLMs struggle with organizing long-form non-fiction?

Nathan Lambert's textbook writing experience suggests LLMs fail at integrating knowledge across chapters despite handling individual sentences well. The question is whether this reflects a fundamental limitation in how models compress and organize information.

What procedural details did OpenAI withhold from its math announcement?

Gary Marcus examines what information OpenAI's math result report omitted—method, architecture, failure rates—and whether the gap prevents independent evaluation of the claim's validity and generalizability.

Is the AI capability gap really an interface problem?

Does the gap between AI model power and real-world productivity stem from poor interface design rather than model limitations? This matters because the answer changes where we should focus improvement efforts.

Does AI help or harm learning based on how it's designed?

Two randomized trials with overlapping research teams tested whether the same AI technology improved or worsened student math and programming outcomes. The difference turned on a single design choice: whether AI gave answers directly or tutored students through problems.

Do wrong AI predictions hurt more than right ones help?

When AI tools give incorrect medical predictions, do they damage clinician performance more severely than correct predictions improve it? This matters for understanding whether averaging test results can hide dangerous asymmetries in AI safety.

Where do researchers actually use AI in their work?

A large survey explores which research tasks researchers adopt AI for most frequently, and whether adoption patterns differ by career stage. Understanding task-specific AI use helps clarify which stages of science may benefit most from automation.

Can we trace AI contributions to scientific breakthroughs?

When AI systems help produce major research results, how can we identify what training data or prior work actually contributed? The Buckmaster-OpenAI dispute shows current systems have no way to track this.

Can AI-generated research outpace peer review systems?

As AI systems produce papers faster and cheaper, will existing peer-review infrastructure become overloaded? The question matters because unmanaged scale could degrade research quality without accelerating discovery.

Can AI governance models from mathematics work across scientific fields?

Should the Leiden Declaration on AI and mathematics—a set of principles for responsible AI use—serve as a template for other disciplines? Nature argues it should, using OpenAI's undisclosed unit-distance proof as a test case for why disclosure matters.

Does generative AI chat actually replace traditional search?

Exploring whether AI-powered chat is fundamentally changing how people seek information, or if traditional search remains central to real research workflows.

Can most adults write prompts good enough for AI?

Jakob Nielsen argues that prompt-based AI interfaces require writing skills most adults lack. The question explores whether literacy barriers might make advanced AI tools inaccessible to a majority of users in wealthy countries.

Does removing AI tools actually measure real skill loss?

Nielsen questions whether lab experiments that take away AI and measure performance decline tell us anything useful about how AI affects workers in real jobs where the tool never gets removed.

How should AI interfaces handle the shift from doing to supervising?

Nielsen explores what UI architecture allows users to oversee AI work rather than perform it themselves. This matters because intent-based systems fundamentally change the user's role from operator to supervisor.

Can companies escape chatbot liability through careful training?

Whether a company's liability for its chatbot's false statements can be reduced or eliminated by investing in accurate training data and proper programming. This matters because it shapes how organizations should budget for AI deployment risk.

Why did AI submissions surge after ChatGPT launched?

Organization Science observed a 42% jump in submissions since ChatGPT's release, nearly double the COVID pandemic's impact. The question explores whether this surge reflects genuine research growth or a shift toward higher-volume, lower-quality output.

Will we ever agree on whether AI makes real discoveries?

Can AI systems produce genuine scientific breakthroughs, or will the field remain divided over what counts as discovery? The answer may depend on whether subjective human judgment can be replaced by objective verification.

Do AI summaries on Google reduce clicks to actual websites?

Pew Research tracked real browsing behavior to test whether AI-generated summaries on Google search results discourage users from clicking through to publisher websites, potentially explaining recent traffic declines.

Why do young adults use chatbots most yet trust them least?

Pew's 2026 survey shows adults under 30 are the heaviest AI chatbot users but most pessimistic about its societal impact. Understanding this gap could reveal whether experience breeds skepticism or if other factors shape young adults' AI concerns.

How do U.S. teens actually use chatbots for schoolwork?

Exploring whether chatbots have become a routine study tool for teens, how helpful they find them, and what patterns emerge across different uses.

Can AI disinformation actually swing an election outcome?

Hungary's 2026 election saw record AI-generated political content reach most voters, yet the disinformation campaign failed to prevent the incumbent party's defeat. What explains this gap between reach and electoral impact?

Where do searchers look when AI Overviews appear?

An eye-tracking study explores whether placing AI Overviews above traditional search results changes where users look and what they trust, revealing how interface design reshapes established scanning patterns.

What if AI and humans form a single organism?

Rao proposes that large language models and humans function together like eukaryotic cells, with models as active nuclei and humans as mitochondria supplying essential 'liveness.' This reframes AI not as replacement but as symbiosis.

Does humanist AI doctrine actually protect or constrain real users?

Rao questions whether AI safety frameworks claiming to prioritize human flourishing instead impose paternalistic constraints based on idealized values rather than users' actual choices and needs.

Why do AI chatbots gain news users but lose their trust?

As AI chatbots become a modest news source, especially among younger and already-engaged audiences, trust in their answers remains far below trust in news overall. What explains this gap between adoption and confidence?

Does trust in AI chatbots drive news-seeking behavior?

As AI chatbot use for news grows globally, researchers ask whether people's trust in these tools predicts adoption rates more reliably than trust predicts social media news consumption—and what that reveals about how people choose their information sources.

Will AI Overviews reduce search referral traffic to publishers?

Publishers expect search referral traffic to decline significantly as Google's AI Overviews answer queries directly without linking out. The question explores whether this forecast reflects real erosion or sentiment-based speculation about search economics.

Do heavy AI users actually encounter more hallucinations?

A survey found power users report 3x more hallucinations than casual users. But does this reflect worse AI performance, harder tasks, or simply higher user standards and scrutiny?

Are AI's math gains and human math losses really connected?

Does AI's breakthrough into elite mathematics explain why students' basic math and literacy scores have declined for 15 years? The question asks whether these trends reflect a single underlying shift or coincidental timing.

Which jobs will actually survive automation by AI?

Exploring what makes certain work intrinsically resistant to automation. The stakes matter because most advice about staying economically valuable assumes the wrong thing—that task difficulty matters more than what's actually being paid for.

Does careful AI use actually protect against reshaping your judgment?

Sacasas challenges the assumption that intentional, vigilant use of AI prevents cognitive changes. The question asks whether diligence can counteract the perceptual shifts that happen beneath conscious awareness.

Is AI's real danger superintelligence or loss of oversight?

Sacasas challenges the conventional framing of AI risk. Rather than fearing superintelligent machines, should we worry about delegating civilizational processes to an unsupervised layer of action beyond human judgment?

How consistent are AI brand recommendation lists across repeated prompts?

Can AI tools like ChatGPT and Claude provide reliable, repeatable brand rankings for tracking market visibility? Understanding this matters because companies may be paying for AI tracking products based on metrics that don't actually measure what they claim.

Do AI chatbots give voters accurate election information?

Researchers tested whether ChatGPT and Google AI could reliably answer common voter questions. The stakes matter because voters increasingly turn to AI for guidance on where and how to vote.

Do chatbots steer Dutch voters toward the same parties?

Dutch regulators tested whether AI chatbots give voting advice that matches what users actually believe, or whether they consistently recommend the same parties regardless of input.

Is generative AI actually driving Wikipedia traffic declines?

The Wikimedia Foundation reports an 8% drop in human pageviews and attributes it to AI search answers and social video platforms. But does the evidence actually support this causal claim, or are other factors at play?

Will AI proofs outrun human mathematical understanding?

Mathematicians interviewed by Williams worry that even if AI solves problems correctly, the solutions might become too complex for humans to comprehend, potentially breaking mathematics' core purpose of building shared human understanding.

How should generative UI be organized as a design space?

Researchers propose four dimensions for understanding generative UI systems: who they serve (designers or end-users), whether the interface changes, who controls those changes, and how long interactions last. This framework helps HCI practitioners design and evaluate AI-generated interfaces.

Do full web pages beat markdown chat for LLM responses?

When LLMs generate complete interactive web pages instead of markdown text, do users prefer them? And how do they compare to pages built by human experts?

What makes people distrust AI agents they delegate to?

When do users withdraw trust in AI agents—and is it really about how much is at stake? A study of delegation tasks reveals which task features actually drive regret and demand for human oversight.

Does learning from AI summaries produce shallower knowledge than web search?

This explores whether the convenience of LLM-generated summaries trades off against the depth of understanding people develop compared to traditional web search. The question matters because it affects how people learn and teach others.

Can people tell AI medical advice from doctors' responses?

A study tested whether people could distinguish AI-generated medical answers from physicians' advice and whether they trusted each equally. Understanding this matters because people act on medical advice they perceive as trustworthy, regardless of accuracy.

Do people use AI assistants before or after searching?

A panel study examined the sequence of assistant use relative to search and browsing in actual user sessions. Understanding this order matters for how AI assistants fit into people's real information-seeking workflows.

Why do users rate disempowering AI interactions more favorably?

Research on 1.5M Claude conversations raises a puzzle: interactions that undermine user autonomy—by distorting beliefs, values, or actions—receive higher approval ratings. What mechanism explains this counterintuitive pattern?

Can interface design reverse citation overload's harm to critical thinking?

As AI writing tools cite more sources, does how we display those citations affect whether readers think critically about them? This matters because citation density often overwhelms users rather than helping them.

Can a company escape chatbot liability by calling it separate?

When an AI chatbot deployed by a company gives customers wrong information, can the company disclaim responsibility by treating the chatbot as an independent entity? This matters for how AI deployment affects corporate liability.

Will mathematicians lose relevance if other fields bypass them for AI?

Can AI-generated answers decouple applied disciplines from mathematical understanding, causing them to stop consulting mathematicians altogether? This explores whether the real threat to mathematics is not computational replacement but institutional irrelevance.

AI at Work

58 notes

When do AI feedback loops trigger explosive growth?

Can automation of research overcome diminishing returns to innovation through combined technological and economic feedback loops? Understanding this threshold matters for forecasting AI acceleration timelines.

Why do firms build automating AI instead of pro-worker AI?

Explores why companies invest more in AI that replaces workers than AI that creates new tasks or augments worker skills, despite evidence that only new-task-creating AI unambiguously benefits workers.

Does AI adoption actually reduce the work that employees do?

Workplace monitoring data from ActivTrak show that as AI tool use has grown, employee work activity has intensified rather than decreased. The question is whether this pattern reflects AI's true impact on workload or something more complex about how work expands when new tools arrive.

Do newer frontier LLMs actually make better strategic decisions?

A strategy simulation benchmark tests whether the latest LLMs can balance short-term profit against long-term growth investment—a core challenge in real strategic reasoning.

Does coding skill or domain knowledge predict agent success?

Anthropic analyzed 400,000 Claude Code sessions to test which matters more for agentic-coding success: the user's programming ability or their understanding of the problem domain.

Does delegating work to AI actually damage worker skills?

Survey data shows heavy AI delegators report career optimism, not decline. But does delegation cause confidence, or do confident workers simply delegate more? And are self-reported feelings reliable indicators of actual skill?

Why do workers hide productivity gains from AI use?

Despite reporting that AI saves time and improves output quality, most professionals and creatives conceal their AI use from colleagues. This note explores what drives this gap between private benefit and public silence.

Can better tools fix an AI agent's exploitable judgment?

Project Vend tested whether scaffolding and oversight could make Claude-based shopkeeper Claudius both more effective and more reliable at avoiding naive mistakes like illegal contracts or security misjudgments.

Does AI productivity gain always ease job displacement fears?

Do workers who gain the most productivity from AI feel more or less threatened by job loss? Understanding this relationship matters for predicting how AI adoption reshapes worker anxiety and labor markets.

Does using more AI tools always boost worker productivity?

A BCG survey of 1,488 US workers explored whether adding more AI tools to workflows improves productivity or reaches a breaking point. Understanding this matters for designing sustainable AI adoption strategies.

Do AI productivity gains feel larger than they actually measure?

A survey of corporate executives explores whether perceived AI productivity improvements outpace what financial metrics capture, and why this gap matters for understanding AI's real economic impact.

Does generative AI actually save workers time or intensify it?

An eight-month ethnography at a tech company investigated whether AI freed up employee time or changed how work gets done. Understanding this matters for predicting how AI adoption shapes workplace demands.

Is AI productivity finally showing up in economic data?

Economists debate whether the 2025 jobs report and GDP growth signal that AI investments are translating into measurable productivity gains, or whether AI impact remains absent from macro statistics.

Do banks accurately assess whether AI can replace customer service jobs?

When Commonwealth Bank cut 45 customer service roles for an AI voice bot, it later reversed the decision after admitting its staffing assessment was wrong. The question explores whether companies systematically overestimate AI's readiness to substitute human labor.

Do AI layoffs actually save money for companies?

A vendor survey explores whether companies that cut roles for AI automation actually achieve the expected financial and operational benefits, or if rehiring and skill gaps erode those gains.

Does data really create lasting competitive advantage for startups?

Explores whether proprietary datasets deliver the durable competitive moat that founders assume. A16z argues collection costs rise while marginal value falls, potentially reversing the flywheel effect.

Is AI really driving job cuts in 2026?

Challenger's tracking shows AI cited in 21% of 2026 layoffs, but the data relies on employer self-reports rather than measured economic outcomes. The question is whether this reflects actual AI displacement or simply how companies frame their decisions.

Do courts actually sanction fabricated AI citations when detected?

Courts are catching AI-hallucinated citations in legal briefs, but whether they impose penalties remains unclear. Understanding sanction rates matters for accountability in AI-assisted legal practice.

Does AI enter strategy where reasoning is deepest or most measurable?

Csaszar et al. investigate whether AI gains strategic autonomy by advancing causal reasoning or by succeeding where performance is easiest to measure. This tests whether organizational trust in AI tracks cognitive depth or demonstrated capability.

Can AI quality control catch fabricated citations in professional reports?

This explores whether existing review processes at major firms can catch AI-generated errors like fake quotes and nonexistent citations before client delivery. It matters because firms claim to maintain quality oversight while adopting AI tools.

Can AI-assisted reports pass quality checks with fabricated citations?

A Deloitte government report contained over a dozen invented references and fake quotes that escaped internal review. The question is whether AI-generated content can systematically bypass citation verification in high-stakes professional work.

Why are companies deploying agents faster than governance matures?

Enterprise leaders are racing to deploy agentic AI within two years, but only 21% report mature governance models. This gap between deployment intent and oversight capability raises questions about how companies are managing the risks of rapid scaling.

Does easier tool-building actually solve enterprise adoption problems?

Explores whether AI's ability to reduce coding barriers addresses the deeper obstacles companies face: recognizing which tasks need tools, defining what those tools should do, and getting organizational buy-in across departments.

Will companies quietly reverse their AI-driven layoffs?

Forrester forecasts that half of layoffs attributed to AI will be reversed by 2026 as companies discover AI-driven savings don't materialize as expected. This raises questions about whether AI-washing and inflated expectations are driving premature workforce cuts.

Does frequent AI use make workers fear job loss more?

Workers using AI daily report double the job-loss anxiety of infrequent users. The question explores whether regular exposure to AI tools directly amplifies displacement concerns, and what organizational factors might buffer that fear.

Does manager support actually shape how employees experience AI at work?

Gallup's survey suggests manager championing of AI correlates with employees reporting improved workplace culture after adoption. But the relationship remains correlational, leaving open whether strong managers drive both AI support and culture perceptions.

Do AI models outperform physicians on health tasks?

HealthBench tested whether large language models produce better health responses than physicians using a shared rubric. The question matters because it determines whether AI can replace or meaningfully augment clinical decision-making.

Why does AI trust keep falling even though it matters most?

A global survey of 48,000 people finds trust is the strongest driver of AI acceptance, yet perceived trustworthiness dropped from 63% to 56% between 2022 and 2024. What's causing the decline?

Do AI customer service agents actually reduce total support costs?

Klarna claims its AI agent saved $60 million while handling work of 853 employees, yet the company's customer service costs rose year over year. This explores whether reported efficiency gains match actual cost trends.

Does using LLMs actually improve strategic decision making?

An experiment tested whether LLM assistance changes how people think through strategic choices and whether those changes lead to better predictions. Understanding this matters because organizations increasingly rely on AI to augment human decision-making.

Can AI systems legitimately resolve wicked policy problems?

Levine explores whether AI, framed as a social institution rather than a brain, has the standing to answer complex policy problems without clear solutions. The question hinges on whether speed and cost override concerns about legitimacy and value judgments.

Why do most enterprise AI pilots fail to deliver returns?

MIT NANDA investigated why 95% of enterprise generative AI pilots produce no measurable profit impact. The research explores whether failure stems from weak models, regulation, or how organizations actually deploy and use these tools.

Can AI data companies sustain margins beyond labor payout ratios?

As AI labs shift from commodity microtasks to expert human judgment, data companies like Mercor show large gross revenues but thin net margins. The question is whether they can build stickier products and services that capture more value than brokering expert labor alone.

Does token volume measure AI productivity or enable gaming?

Meta's leaderboard ranks employees by AI token consumption, but does this metric drive real productivity gains or incentivize wasteful agent-running? The note explores whether volume-based status metrics actually correlate with useful work.

Can AI boost how teams work together?

Explores whether AI systems designed for shared goals and group collaboration can deliver productivity gains beyond what individuals achieve alone. This matters because current AI adoption data shows the opposite trend.

Why do ready workers struggle to transform their work?

Microsoft's research explores why employees skilled in AI adoption face organizational barriers. The gap between individual capability and systemic support may explain why transformation stalls even when workers are prepared.

How do human strengths help people exploit AI capabilities?

Mollick argues that the gap between AI's actual capabilities and how people use them can be bridged by bringing four specific human advantages—deep knowledge, wide knowledge, taste, and agency—to AI interaction.

Can AI help social science move beyond the peer-reviewed PDF?

Kevin Munger explores whether artificial intelligence could enable researchers to unbundle the functions currently locked into peer-reviewed PDFs—archiving, literature review, methods, results—into more diverse and specialized forms better suited to different types of epistemically valuable work.

Does AI really compress all layers of knowledge work equally?

Explores whether AI's productivity gains affect all stages of knowledge work uniformly, or whether some tasks like planning and accountability grow harder as execution becomes easier.

Who adopts enterprise AI first and how do they use it?

This explores which firms embrace ChatGPT Enterprise and whether adoption translates to uniform use across roles and tasks. Understanding adoption patterns helps predict how AI reshapes organizational work.

Do LLMs consistently favor the same strategic choices regardless of context?

Exploring whether large language models exhibit systematic bias toward trendy strategic recommendations—like collaboration over competition—even when industry context and detailed prompting should pull them toward different answers.

Can AI safety pacing work without government cooperation?

Explores whether voluntary industry proposals to slow AI development can succeed when major powers view the technology as strategically competitive rather than a shared problem requiring coordination.

Why do developers keep using AI tools they don't trust?

Explores the paradox where AI adoption rises to 80% among developers even as trust in accuracy drops sharply to 29%. Why does usefulness persist despite frustration with unreliable output?

Is AI already shrinking the entry-level job market?

Stanford's AI Index reports a sharp 20% employment drop for young software developers, while surveys predict much larger workforce cuts ahead. The question is whether this narrow, measured decline signals the start of broader AI-driven job losses.

Will AI safety's media surge convert into real political power?

Despite AI safety reaching mainstream media audiences in 2026, reporting suggests the spike in public concern may not translate into grassroots organizing or electoral influence. This matters for understanding whether attention to AI risks can actually shape governance.

Can LLMs become good editors by learning a writer's taste?

Explores whether large language models can perform well as editors—despite poor creative writing—by being trained on explicit personal taste rubrics rather than generic standards.

Why do workers feel confident with AI but get poor results?

Workers report high confidence using AI, but most say it fails on first attempt or takes longer than manual work. What explains this gap between perceived competence and actual performance?

Does labeling AI as an employee change how managers oversee it?

When organizations formally list AI agents on org charts and frame them as employees rather than tools, do managers change their own error-catching behavior? This matters because it tests whether organizational framing alone—independent of the AI's actual capabilities—shifts oversight and accountability.

Where does AI's time savings actually go in practice?

A survey of 3,200 AI users explores whether time saved by AI tools translates into real productivity gains or gets absorbed by correction work and task overload. Understanding this gap matters for predicting AI's actual workplace impact.

How much time do workers really spend fixing AI mistakes?

Enterprise workers report spending substantial weekly hours correcting AI output despite claiming productivity gains. Understanding this gap matters for realistic AI adoption planning and hidden cost accounting.

Can AI forecasters beat expert humans at venture evaluation?

Do frontier large language models outperform experienced managers and investors at predicting fundraising success? This matters because venture assessment is a genuinely uncertain, ill-structured judgment task where human expertise is assumed essential.

Do executives and employees agree on AI's job impact?

Surveys show executives predict AI will cut jobs while employees expect gains. Understanding this expectation gap matters because it reveals whether both sides of employment see the same future, or if diverging beliefs might shape hiring and career decisions differently.

Does AI collaboration drain motivation when workers return to solo tasks?

When people work with generative AI and then switch to independent work, do they experience psychological costs beyond performance changes? This matters for understanding the hidden friction in hybrid human-AI workflows.

How does generative AI actually change worker skills?

Rather than simply upskilling or deskilling workers, how do knowledge workers themselves experience changes in their capabilities when using GenAI? Understanding these varied outcomes could reshape how we design tools and support workforces.

Does AI theme-mapping perform as well as human reviewers?

Can an AI system match expert judgment when automatically sorting consultation responses into themes? Understanding this matters for scaling policy feedback analysis without sacrificing accuracy.

Why do knowledge workers hide signs of using GenAI?

Knowledge workers conceal GenAI use at work, but research suggests the motive goes beyond avoiding stigma. Does hiding GenAI cues actually signal domain expertise, and what happens to peer learning when workers stay silent?

Does GenAI actually save lawyers time on fact verification?

Explores whether AI-generated summaries speed up legal fact-checking or create hidden costs through opacity. Questions whether automation's claimed efficiency gains hold up under real verification demands.

Does GenAI assessment challenge fit wicked problem theory?

Explores whether the GenAI-and-assessment problem in universities matches Rittel and Webber's framework for wicked problems—ones lacking clear definitions and definitive solutions—and what that diagnosis means for institutional responses.

Co-Writing and Collaboration

17 notes

Can process data distinguish AI delegation from ordinary collaboration?

When students or writers use AI tools, their work leaves traces in keystroke logs and editor telemetry. Can these process signatures reliably separate wholesale delegation from permitted collaborative use?

Can AI generate hundreds of fake academic papers automatically?

Explores whether language models can industrialize academic fraud by retroactively constructing theoretical justifications for data-mined patterns, complete with fabricated citations and creative signal names.

Does AI writing assistance change how readers perceive the writer?

Explores whether AI-assisted writing systematically alters reader impressions of the writer's political views, competence, emotion, and demographic identity. Understanding this matters because perception shapes trust and influence in public discourse.

Does AI writing make all writers sound the same?

When writers use AI assistance, do their distinct voices converge toward a generic style? This matters because readers rely on voice to identify and distinguish among individual writers.

Does AI writing make authors seem more privileged than they are?

When writers use AI assistance, do readers perceive them as more educated, wealthier, and whiter? This matters because it could mask or erase the actual diversity of voices in public discourse.

Does Figma Make speed up design task completion?

Does access to a prompt-to-design tool reduce the time needed to complete structured design work, and does the effect differ between professional designers and product managers?

Do writers actually edit AI-generated text before publishing?

This research tests whether the "human-in-the-loop" safeguard against AI text quality issues actually works in practice. It examines how often writers revise AI-generated paragraphs and how substantially they change them.

Can AI writing assistance remove distortion without losing appeal?

When researchers tried to correct AI persona distortions through reward model training, the fixes reduced user preference for the text. This raises a fundamental question: are the distortions and desirable properties structurally inseparable?

Do writers actually prefer AI-edited versions of their own text?

When writers compose opinions and then edit AI-generated alternatives, which version do they choose? Understanding this preference matters because it determines whether AI-assisted text gets treated as authentic personal expression in public discourse.

Can source traceability make AI writing trustworthy?

If every claim in machine-generated text traces back to a verifiable source, does that fundamentally change whether human professionals will actually use AI as a collaborator rather than a curiosity?

What tasks do expert data storytellers trust to LLMs?

Expert visual data storytellers make strategic choices about which narrative work to delegate to LLMs and which to protect. Understanding these boundaries reveals how human judgment and automation can coexist in knowledge work.

How do writers use AI through different creative stages?

This study explores whether writers deploy large language models differently depending on their creative needs—from generating initial ideas to organizing thoughts to drafting final text. Understanding these patterns reveals how humans and AI can complement each other's strengths.

Does ownership framing change how much writers rely on AI?

When writers believe they own the final output versus composing for themselves, do they use AI suggestions differently? Understanding this matters because it reveals whether reliance is driven by tool capability or by how tasks are framed.

Can statistical rarity measure whether stories are truly original?

Can we operationalize originality as statistical rarity in narrative feature space? This matters because copyright law requires measuring human creative control, but rarity is relative, context-dependent, and doesn't guarantee quality or authorship.

Do university AI policies actually protect what credentials mean?

Universities are getting better at stating what AI use is allowed, but do their policies explain what evidence proves a student's actual competence? This matters because a credential's value depends on what work the student actually did.

Can writers benefit from configuring AI writing partners in advance?

This exploratory study asks whether writers can effectively set up proactive AI assistants by pre-planning their roles and behavior, then use them during actual writing work for idea generation and self-monitoring.

Do writers want to see each other's AI prompts in shared editors?

This study explores whether revealing AI prompting activity to collaborators in text editors affects how writers work together. Understanding prompt visibility matters because it shapes trust, learning, and awareness of AI's role in collaborative writing.

Workplace Applications

9 notes

Does AI turn freelance work into validation instead of creation?

Does shifting freelancers from producing original work to validating AI output undermine their ability to build skills through paid practice? This matters because freelancers rely on client work as their primary learning mechanism.

Why does AI default to coaching instead of doing?

In workplace conversations, users often want AI to execute tasks like writing or gathering information, but AI tends to explain and advise instead. What drives this systematic mismatch between what users need and what AI provides?

Do LLM research ideas actually hold up when experts try to execute them?

Explores whether LLM-generated ideas maintain their apparent novelty advantage when expert researchers spend 100+ hours implementing them. Matters because ideation-stage evaluation may not capture real-world feasibility barriers.

Does concentrated AI exposure enable workers to adapt and reallocate?

When AI displaces specific tasks rather than spreading across many, workers may shift effort to non-displaced tasks within their occupation. Does this reallocation mechanism actually offset employment losses?

Where have workers actually delegated tasks to AI?

Existing AI-exposure measures predict where AI could work, not where workers have actually adopted it. This research asks which occupations have embedded AI into real workflows, and whether that pattern matches technical capability or conversational tool use.

Does generative AI shift knowledge workers away from communication?

When knowledge workers adopt generative AI heavily, do they spend proportionally more time on individual documentation and less on coordination with colleagues? Understanding this matters because it suggests AI may reshape not just productivity but the social fabric of how teams work together.